Field of the Invention
The present invention relates to a solid-state image sensor and a camera.
Description of the Related Art
Solid-state image sensors include, for example, pixel arrays in which a plurality of pixels are arrayed. Some solid-state image sensors include focus detection pixels for performing focus detection arranged in the pixel arrays. Each focus detection pixel can include, for example, a pair of charge accumulation regions, a floating diffusion, and two transfer transistors each transferring a charge accumulated in the pair of charge accumulation regions by photoelectric conversion to the floating diffusion.
More specifically, the charge corresponding to a part of light that has passed through an optical element (for example, a microlens or the like) is accumulated in one charge accumulation region and a charge corresponding to another part of the light is accumulated in the other charge accumulation region. Then, a signal (first signal) corresponding to an accumulated charge amount in one charge accumulation region and a signal (second signal) corresponding to an accumulated charge amount in the other charge accumulation region are used to perform focus detection based on a phase-difference detection method. The first signal and the second signal are added to be processed as a pixel signal, and are also used to form image data.
If the charges accumulated in a charge accumulation region are saturated because an amount of light entering a pixel is sufficiently large, charges further generated from that state by photoelectric conversion (to be referred to as “excess charges” hereinafter) may leak into another pixel adjacent to the pixel.
Japanese Patent Laid-Open No. 2013-84742 has disclosed a structure in which a potential barrier between one charge accumulation region and another charge accumulation region of a pair of charge accumulation regions in each pixel is lower than a potential barrier between adjacent pixels. According to this structure, if charges accumulated in one charge accumulation region are saturated, excess charges can be received in the other charge accumulation region. It is therefore possible to prevent leakage of the excess charges into the adjacent pixels and a decrease in image quality.
According to this structure, however, because the potential barrier between one charge accumulation region and the other charge accumulation region of the pair of charge accumulation regions has been lowered, some of the excess charges may leak into a floating diffusion when the excess charges move from the one charge accumulation region to the other charge accumulation region.